function score = bayesian_score_CPD(CPD, local_ev)% bayesian_score_CPD Compute the Bayesian score of a tabular CPD using uniform Dirichlet prior% score = bayesian_score_CPD(CPD, local_ev)%% The Bayesian score is the log marginal likelihoodif iscell(local_ev) data = num2cell(local_ev);else data =	local_ev;endscore = dirichlet_score_family(compute_counts(data, CPD.sizes));